A diagnostic model of polycystic ovary syndrome was constructed based on the genes associated with pyroptosis

Background: Polycystic ovary syndrome (PCOS) is a complex disease, and the underlying mechanisms remain unclear. It has been suggested that genes involved in pyroptosis may play a regulatory role in PCOS. However, the exact contribution of pyroptosis to PCOS is not fully understood. Methods To investigate this, we obtained three mRNA expression profiles from the Gene Expression Synthesis (GEO) database and analyzed the differential expression of pyroptosis-related genes (PRGs) between PCOS patients and normal individuals. We employed four machine learning algorithms (GLM, RF, SVM, and XGB) to identify disease signature genes. Results A predictive model and a nomogram were developed based on PRGs to accurately predict PCOS. The XGB method demonstrated the highest accuracy in validating the model using two independent datasets, which was further supported by decision curve analysis. Consensus clustering revealed two distinct subgroups within PCOS cases, with Cluster2 exhibiting higher immune infiltration compared to Cluster1. Differential expression analysis identified DEGs between the two subtypes, and pathway enrichment analysis was conducted on the model genes. Conclusion This study provides preliminary insights into the association between PCOS and pyroptosis, and presents a precise predictive model for PCOS.